Issue metadata
Sign in to add a comment
|
1%-14.4% regression in system_health.memory_mobile at 461404:461499 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Apr 5 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983144851082653120
,
Apr 6 2017
=== Auto-CCing suspected CL author jwolfe@igalia.com === Hi jwolfe@igalia.com, the bisect results pointed to your CL, please take a look at the results. === BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Josh Wolfe Commit : 2474b73472bbe5313ed043b348e93b96167423dd Date : Mon Apr 03 08:40:26 2017 Subject: Convert String.prototype.normalize from JS to CPP builtin Bisect Details Configuration: android_webview_arm64_aosp_perf_bisect Benchmark : system_health.memory_mobile Metric : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/blank_about/blank_about_blank Change : 14.24% | 2745544.0 -> 3136408.0 Revision Result N chromium@461410 2745544 +- 0.0 6 good chromium@461412 2745544 +- 0.0 6 good chromium@461412,v8@8af394d6d3 2745544 +- 0.0 6 good chromium@461412,v8@2474b73472 3136584 +- 329.266 6 bad <-- chromium@461412,v8@5833e77744 3136555 +- 203.227 6 bad chromium@461413 3136437 +- 321.331 6 bad chromium@461416 3136496 +- 248.902 6 bad chromium@461421 3136555 +- 321.331 6 bad chromium@461431 3136408 +- 329.266 6 bad Please refer to the following doc on diagnosing memory regressions: https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md To Run This Test src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=blank.about.blank system_health.memory_mobile Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8983144851082653120 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5809280818610176 | O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq | X | for more information addressing perf regression bugs. For feedback, | / \ | file a bug with component Speed>Bisection. Thank you!
,
Apr 12 2017
,
Apr 12 2017
The blamed patch might have caused a tiny increase in memory usage due to the addition of four string constants. But beyond that, prior patches which removed JS builtins code have led to bumpy, observable changes on certain benchmarks which went away based on other unrelated-looking changes. We have a broader project goal of eliminating all of this code, and the blamed patch does this move in a straightforward way. Recently, there was a significant regression in this metric many times greater than the regression blamed on this patch. I think the right action to take here would be to keep an eye on this dashboard and see if the regression blamed on this patch is still there when the bigger, unrelated regression is resolved.
,
Apr 12 2017
I also don't trust the bisect, fwiw (marked it as "wrong" from the link above).
,
Jul 27 2017
Explictly assigning. A CL you landed tripped one of the speed metrics we measure in the lab. If this is the first time this has happened to one of your CLs, or if it's been a while, please read: https://chromium.googlesource.com/chromium/src/+/master/docs/speed/addressing_performance_regressions.md We're looking for one of the following: 1. Justification via explanation 2. Plan to revert or fix 3. Angry rage throwing of equipment at my head Just be aware that I'm trained in trumpet playing and First Aid and am not afraid to use it. Note: This was a bulk edit message and not very personal.
,
Aug 1 2017
You can find such a justification above. I don't think we need a revert or fix.
,
Aug 9 2017
|
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by pmeenan@chromium.org
, Apr 5 2017